Looped quality improvement project of consent forms in patients admitted with hip fracture

Abstract

Introduction

Neck of femur (NOF) fractures constitute a significant proportion of the workload of orthopaedic departments, with approximately 65 000 admissions across the UK annually. There are various surgical procedures available to repair NOF fractures and each is associated with numerous intraoperative and perioperative risks. Informed consent is an essential part of the workup when planning to undertake any surgical procedure, whereby the risks and benefits of surgery are discussed with the patient, so they are able to make decisions about their care. It is therefore important that consent forms are sufficiently detailed and accurate to ensure consent is fully informed.

Method

We performed a multi-loop audit of consent forms for patients admitted to the Royal Oldham Hospital (United Kingdom) with NOF fracture. The frequency that different risks were included on consent forms was compared to the British Orthopaedic Association (BOA)-produced standardised list of risks for NOF fracture procedures.

Results

We found that on each consent form an average of only 56.6 % of risks were included before intervention. Some significant risks such as mortality were frequently missed. Following this, we implemented changes to the consent process including the use of stickers to use on consent forms and team education sessions. We subsequently performed a second audit cycle and found that compliance had increased to 87.9 %.

Conclusion

Overall, there was poor documentation of surgical risks on consent forms, however this was improved with basic interventions. A further audit cycle should be recommended to investigate if this improvement is sustained.

1. Introduction

Informed consent is an essential part of medical practice when planning to undertake any surgical procedure. The Montgomery case¹ highlighted the importance of ensuring that patients are provided with all the information they require to make an informed decision. This landmark case found that a patient was not informed of all options available to them or all the risks involved when planning the birth of their child, following which the baby sustained shoulder dystocia. This subsequently resulted in a change to GMC guidance on taking consent,² which moved from a primarily ‘doctor-focussed’ to ‘patient-focussed’ framework, whereby the patient is supported in making their own decisions about their health and treatment. Whilst this has affected clinical practise, it has also affected medico-legal activity in the UK. The claims made against the NHS in relation to consent have seen an increase from 3.61 % of all claims in 2011–15 (£14.07 million), compared to 7.91 % (£25.18 million) in 2016–2020.³

Neck of femur (NOF) fractures make up a significant proportion of the workload of orthopaedic departments, with over 70 000 admissions occurring in hospitals in the UK in 2022 alone.⁴ NOF fractures can occur as fragility fractures in the elderly or as high velocity fractures affecting younger and middle-aged individuals. Since the elderly are a generally comorbid population it is an injury associated with high morbidity and mortality. 1-Year mortality is reported to be 16–25 %, with a three to four times risk of mortality post-surgery compared to the average population.^5,6

Surgical treatment options for NOF fractures can include dynamic hip screw (DHS), intramedullary (IM) nail, cannulated cancellous (CC) screw, total hip replacement (THR) and hemiarthroplasty. All procedures are associated with numerous intraoperative and perioperative risks such as risk of lower limb deep vein thrombosis, pulmonary emboli, peri-prosthetic fracture and damage to local nerves and vessels. The British Orthopaedic Association (BOA) has endorsed a standardised list of risks for each procedure⁷ (available at orthoconsent.com). Because the population undergoing these procedures is largely frail, the risks of surgery are relatively high. Therefore, it is important that these risks are discussed in full of each patient prior to surgery as part of the process of informed consent.

Previous studies looking at consent form accuracy highlight that the consent for NOF fracture surgery is generally poorly performed.^{8, 9, 10, 11, 12, 13, 14} They revealed that surgical risks are not accurately recorded on consent forms, with pertinent and significant procedural risks frequently missed. Probert et al.,¹³ and Shah et al.¹² suggested the use of consent form stickers however did not re-audit their data after the intervention. From our literature review, there has only been one study in the elective setting which audited the introduction of consent stickers for hip arthroplasty.¹⁵

We decided to perform a closed loop quality improvement audit aiming to analyse the effect of interventions such as consent stickers and clinical education to improve consent form compliance for NOF fracture patients in the trauma setting. Our project collected data to analyse the variation and accuracy of procedural risks listed on consent forms for patients undergoing surgery for repair of NOF fractures at the Royal Oldham Hospital (ROH). We measured this against the guidelines endorsed by BOA as our benchmark. The ROH is a district general hospital in the North-West of England, with a busy trauma and orthopaedic department. Following our first cycle of data we introduced several interventions to improve consent form accuracy. There are ethical, moral and legal obligations to ensure informed consent is carried out properly.

2. Methods

We retrospectively reviewed daily on-call take lists to identify patients admitted to the Trauma and Orthopaedics department at the Royal Oldham Hospital, United Kingdom, with a neck of femur fracture. The period assessed for the first audit cycle was between April 2023–June 2023.

Patient records were then accessed for each identified patient with eligibility being assessed using the following exclusion criteria.

• 1)Non-capacitate consent received (consent form 4).
• 2)Consent form not uploaded to scanned notes.
• 3)Fracture managed non-surgically.
• 4)Consent form handwriting illegible.

For eligible patients, the type of surgery performed was determined and then scanned consent forms were reviewed to assess which risks were mentioned. All the data was collated in an electronic database which was pseudonymised to include only the patient's hospital number.

The standard we compared the list of risks on each consent form were to the BOA-endorsed lists of risks for each surgical procedure used for NOF fracture repair (dynamic hip screw, IM nailing, CC screw, total hip replacement and hemiarthroplasty). These guidelines are available at orthoconsent.com. The target was for 100 % of risks to be included on each patient's consent form.

The collated data was then analysed in GraphPad Prism. Descriptive statistics and illustrative graphs were produced as appropriate. Data was analysed in two groups depending on the procedure undertaken: 1) dynamic hip screw, IM nailing or CC screw and 2) total hip replacement/hemiarthroplasty. This was because the risks for the procedures in each group are similar.

In November 2023, we implemented a series of interventions around consent forms for NOF fracture surgery. These included.

• 1)Pre-printed stickers listing all risks for each surgical procedure made available in the trauma office.
• 2)Educational email sent to all members of the orthopaedic medical team.
• 3)Presentation of our first cycle data to the departmental clinical governance meeting

Following this, the effect of these interventions was measured via a second audit cycle using the methods as described above. The period assessed in this cycle was December 2023–January 2024.

3. Results

Over the period analysed in the first audit cycle, a total of 111 patients were referred to the hospital's orthopaedic department with neck of femur fracture. After assessing eligibility, a total of 41 patients who underwent DHS/IM nailing/CC screw were analysed, and 34 patients who underwent THR/hemiarthroplasty. A flowchart of participants included is shown in Fig. 1.

Fig. 1.

Flowchart of patients analysed in the first audit cycle.

3.1. DHS/IM nail/CC screw

For DHS/IM nail/CC screw, a total of 13 procedural risks were assessed according to the BOA-endorsed guidelines. Across the 41 patients analysed, a total of 293 risks were mentioned on the consent forms out of a possible 533. This means the average compliance was 55 %. However, the frequency at which each risk was mentioned was not evenly distributed. Infection, bleeding and blood clots were mentioned on 100 % of forms, whilst avascular necrosis was only mentioned two times and urinary catheter was not mentioned at all. This is summarised in Table 1 and Fig. 2.

Table 1.

Risks included on consent forms for DHS/CC screw/IM nail.

Risk		Freq. mentioned on consent form [%]
1	Infection	41/41 100
2	Bleeding	41/41 100
3	Blood clots	41/41 100
4	Damage to nerves/vessels/other	40/41 [97.6]
5	Scar/altered wound healing	34/41 [82.9]
6	Failure/further surgery	27/41 [65.6]
7	Pain	24/41 [58.5]
8	Mortality/death	20/41 [48.8]
9	Altered leg length	10/41 [24.4]
10	Stiffness	7/41 [17.1]
11	Periprosthetic fracture	6/41 [14.6]
12	Avascular necrosis	2/41 [4.9]
13	Catheter	0/41 0

Fig. 2.

Summary of risks included for each patient for DHS/CC screw/IM nail. Blue = mentioned, white = not mentioned.

3.2. THR/hemiarthroplasty

For THR/Hemiarthroplasty, a total of 14 procedural risks were assessed. Across the 34 patients, a total of 278 risks were mentioned on the consent forms out of a possible 476. Average compliance therefore was 58.4 %. Similarly, to DHS/CC screw/IM nail, there was variation in the frequency different risks were mentioned. Again infection, bleeding and blood clots were always mentioned. Prosthesis loosening, stiffness and urinary catheter were infrequently mentioned. This is shown in Table 2 and Fig. 3.

Table 2.

Risks included on consent forms for THR/Hemiarthroplasty.

Risk		Freq. mentioned on consent form [%]
1	Infection	34/34 100
2	Bleeding	34/34 100
3	Blood clots	34/34 100
4	Damage to nerves/vessels/other	33/34 [97.1]
5	Scar/altered wound healing	28/34 [82.4]
6	Pain	25/34 [73.5]
7	Mortality/death	20/34 [58.8]
8	Joint dislocation	20/34 [58.8]
9	Failure/further surgery	19/34 [55.9]
10	Periprosthetic fracture	11/34 [32.4]
11	Altered leg length	9/34 [26.5]
12	Prosthesis loosening/metal failure	6/34 [17.6]
13	Stiffness	4/34 [11.8]
14	Catheter	1/34 [2.9]

Fig. 3.

Summary of risks included for each patient for THR/Hemiarthroplasty. Blue = mentioned, white = not mentioned.

4. Intervention and re-audit

Following the first audit cycle, we implemented a series of interventions to address consent form variability. We presented our findings at the local clinical governance meeting and sent an educational email to all members of the orthopaedic clinical team highlighting findings of the first audit cycle. We encouraged the use of pre-printed stickers that we produced, which listed the risks for each procedure and can be placed directly on each patient's consent form and can be counter signed by the clinician. After this, a second audit cycle was performed to measure the impact of these interventions.

During the period assessed for the second audit cycle, a total of 80 patients were referred with neck of femur fracture. Of these, 16 patients who underwent DHS/CC screw/IM nail and 26 patients who underwent hemiarthroplasty/THR, were included in the analysis. A flowchart of patients included in the second cycle analysis is shown in Fig. 4.

Fig. 4.

Flowchart of patients included in the second audit cycle.

Following the implementation of our interventions, the overall average number of risks mentioned on the consent forms for patients undergoing DHS/IM nail/CC screw increased to 86.1 %, compared to 55 % before intervention. Similarly for hemiarthroplasty/THR, compliance increased to 89.0 % from 58.4 %. Average compliance split by each individual risk is shown in Fig. 5. For some risks such as prosthesis loosening, stiffness and urinary catheter which were infrequently mentioned before intervention, there was substantial improvement seen in the second cycle.

Fig. 5.

Average percentage compliance for each individual surgical risk across all consent forms for (a) DHS/CC screw/IM nail and (b) hemiarthroplasty/THR. Black circles indicate pre-intervention compliance and blue squares indicate post-intervention compliance.

5. Discussion

We performed a multi-loop audit on the variation and accuracy of consent forms in detailing procedural risks for patients undergoing surgical repair of NOF fractures. Overall compliance was low compared to BOA-endorsed guidelines (55 % and 58.4 % for DHS/IM nail/CC screw and THR/Hemiarthroplasty respectively). We found a large degree of inconsistency between individual procedural risks as well. Perera et al.¹¹ which found “risks that were most frequently documented were infection, blood clots (deep vein thrombosis and pulmonary embolism), and bleeding, with documentation rates of over 90 %“, our study also echoed that some common surgical risks were mentioned on all consent forms, namely infection, bleeding and blood clots. The general trend was that generic risks applicable to all types of surgery were more commonly mentioned, whilst procedure-specific risks were less-frequently mentioned. For example, in the DHS/IM nail/CC screw cohort, peri-prosthetic fracture was mentioned 14.6 % times, avascular necrosis mentioned 4.9 % and altered leg length 24.4 %. This might suggest that there is a lack of familiarity amongst individuals who consent patients for surgery with the specific surgical procedures for NOF fractures. Perera et al.¹¹ have suggested that a reason for this may be that “majority of the consents were obtained by senior house officers (SHO) and core surgical trainees who did not have prior formal orthopaedic training (52.4 %)”. Notably, some serious and significant risks such as mortality were infrequently mentioned. For DHS/IM nail/CC screw, this appeared on less than half of consent forms. It is very important that patients are made aware of these risks before surgery and further education should be undertaken to address this.

A similar study also recommended the use of pre-printed stickers to improve compliance, however no data was collected for a re-audit.¹² One study found that compliance increased to 100 % when clinicians were provided with detailed consenting guidance which included a model consent form to follow.¹⁰ They introduced procedure-specific pre produced consent forms for extra-capsular NOF fracture which may lead to sustained improvement. If successful, these could be expanded to a range of procedures and considered in other hospitals as well.

In the second cycle of our audit, we saw a significant improvement in documentation of risks mentioned on consent forms. This may be due to clinicians choosing to use the consent form stickers as it is a more efficient way of documentation as compared to manually writing out all the risks. It is therefore less likely that clinicians omitted surgical risks whilst gaining informed consent from their patients as they would have a visual reminder when going through the sticker. Although this cannot be discerned from our data, we would suggest further patient focussed research involving surveys or recorded consultations to see whether it has led to better patient understanding of perceived consents.

Despite our data showing improvement in documentation [Fig. 5] it may be short lived, since most junior doctors have rotational contracts which means that regular re-education and awareness about the stickers will be required to ensure this continues at every cohort of new junior doctors. Hence a further audit loop in a year's time would be recommended to explore this. The fact that multiple studies have found deficiencies in the documenting of procedure-specific risks might suggest this is a broader issue affecting multiple hospitals and various surgical specialties. This suggests it is a topic in which teaching could be improved nationally as part of the curriculum for the core surgical training programme.

The authors acknowledge that there are some limitations of the results presented in this study. This study was conducted at a single centre (Royal Oldham Hospital), which limits the generalisability of the findings to other hospitals or regions. Different institutions may have varying patient demographics, clinical workflows, and approaches to the consenting process, which could influence the results. The audit cycles were conducted over relatively short time periods, with the first cycle spanning three months and the second cycle only two months. This limited timeframe may not accurately reflect long-term trends, or the sustainability of the interventions implemented. A key limitation is the focus on documentation compliance rather than patient understanding. While the study evaluated whether procedural risks were recorded, it did not explore whether patients comprehended the risks discussed during the consenting process. The interventions implemented—pre-printed stickers and educational sessions—were effective at improving documentation but did not explore other methods such as video aids, detailed patient information leaflets, or standardised consent forms. These additional strategies may further enhance informed consent. Many consent forms were completed by junior doctors on rotational contracts. This reliance on transient staff may result in variability in the quality of documentation due to differing levels of training and familiarity with orthopaedic procedures.

6. Conclusion

Documentation of surgical risks on consent forms for fractured neck of femur patients is often done poorly. This is particularly pronounced if the risks are to be handwritten by the clinician on the consent forms. However, educating the department and introducing procedure specific stickers to be counter signed on the forms significantly improves compliance to the standard expected. Further auditing is required to understand whether these interventions will lead to lasting change, does documentation of risks translate to better informed consent by the patient and what should be the gold standard practise to safeguard both patients and clinicians in medicolegal cases regarding consent.

Patient consent

This patient did not collect any personal data as it was the review of their consent forms in a retrospective audit.

Author statement

Aniket Bharadwaj: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Resources; Software; Supervision; Validation; Visualization; Roles/Writing - original draft.

Richard Morter: Data curation; Formal analysis; Investigation; Methodology; Roles/Writing - original draft.

Ardeshir Bonshahi: Supervision; Project administration; Resources; Visualization.

All authors have read and approved the final manuscript. We confirm that all authors meet the criteria for authorship as defined by the journal guidelines.

Ethical statement

Ethical approval was not required for this study, as per the NHS Research Ethics Committee tool. This study did not include experiments on human or animal subjects.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.